Summary Cardiopulmonary toxicities following thoracic radiotherapy and PD-1 blocking immunotherapy have a major impact on quality of life and survival. Therefore, there is an unmet need to have early diagnostic test and intervention for fatal toxicities. Our proposal aims to understand immune mechanisms that modulate potentially life-threatening cardiopulmonary toxicities. We propose to study the toxicities in both mouse models and patients by analyzing blood samples. To achieve our goals, we have assembled a collaborative team across three institutions with a group of expert consultants in order to identify biological correlates and therapeutic targets to ameliorate these autoimmune toxicities. Our prior studies showed excessive mortality in mice simultaneously exposed to radiotherapy and PD-1 inhibition, which we show is dependent on both the cytokine IL-17A and the B-cell. Since both Th17/IL-17A and humoral immunity are implicated in autoimmune diseases, we hypothesize that toxicities result from the unchecked adaptive Th17 response due to PD-1 blockade, combined with autoantibodies against heart and lung tissues generated by the pro-inflammatory B- lymphocytes. We will employ mouse models and pharmacological inhibitors to dissect the underlying autoimmune mechanisms and measure key components of Th17 and B-cell response in prospectively collected blood samples from lung cancer patients undergoing combine radiotherapy and immunotherapy. Aim 1: To determine whether IL-17A/Th17 responses mediate the toxicities. We hypothesize that both innate and adaptive immunity contributes to the toxicities through the link of IL-17A. We will generate KO mice unable to produce IL-17A through either neutrophils or CD4 T cells. We expect that the toxicities are attenuated when Th17/IL-17A are blunted in these mouse models. To determine whether IL-17A/Th17 can be used as predictive biomarkers for the toxicities, we will examine dynamic changes of Th17/IL-17A in serum samples from our patients. Aim 2: To determine the role of humoral response in mediating the toxicities. We hypothesize that pro-inflammatory Tbet+ B-lymphocytes drive autoantibody production which results in the toxicities. We will use Tbetflox/flox CD19cre mice as our model, in which mature Tbet+ B cells are absent. This approach will be complemented by pharmacological depletion of B cells using anti-CD20 or by neutralizing autoantibodies with IVIg in wild-type mice. We expect that the toxicities are attenuated in these models. Furthermore, we will test whether pharmacological inhibitors of Th17/IL-17A reduce Tbet+ B cells and autoantibodies. Finally, the rise of autoantibodies in blood will be captured in mice and patients as a surrogate for the toxicities. Our study of basic mechanisms in preclinical models, combined with analysis of patient samples, will lead to novel diagnostics for early detection and improved therapies for severe cardiopulmonary toxicities.